Because over 85% of lung cancer is caused by smoking, we focus on the effect of tobacco policies on lung cancer. We will extend a previously developed macro-simulation model, known as SimSmoke, to predict smoking-attributable lung cancer the effect of tobacco control policies on those deaths. The first major aim of this project is extending SimSrnoke to estimate and predict smoking-attributable lung cancer deaths. The model will be programmed to estimate the number of smoking-attributable lung cancer deaths to smoker and non smokers during the past ten years and predict deaths over the next 35 years, and to distinguish the number of deaths by age, gender and by racial/ethnic group, and distinguish the effects of quantity smoked and smoking prevalence. We will also distinguish the effect on lung cancer of factors other than smoking, other such as other risks and the effects of treatment. We also propose to determine the impact of tobacco control interventions on observed trends in mortality; and to determine if the interventions are having their expected population impact. Specifically, SimSmoke will be used to estimate the number of smoking-attributable deaths in the United States averted as a result of policies implemented in the last ten years, and estimate the number of smoking deaths that have been averted as a result of policies implemented in the last ten years in three states with proactive tobacco control policy. We will also consider the potential impact of policies in the future. In examining the effect of tobacco control policies, we will distinguish their effects on smoker and non-smoker deaths, their effects by age, gender, and racial/ethnic group, and their effects in terms of quantity reduction and smoking cessation. We will also add a new module to examine the effect of new tobacco products (low tar and cigarettes without certain additives) and non-tobacco products (inhalers) which may reduce lung cancer risk, and a module to consider how improved lung cancer detection and treatment may reduce smoking-attributable lung cancer deaths, and how they might be coordinated with tobacco control policies. A final goal of this project will be to critically examine the methods that are traditionally used to project lung cancer deaths, and determine how the estimates depend on the sensitivity to key parameters.